Chief Scientist in the Measurement Science group, with core competence in modelling and data analysis both on sensor and system level. Kocbach combines system level knowledge with deep knowledge within physics and physical sensor principles for modeling and data analysis on various levels. Research tasks range from sensor design and analysis of sensor data to modeling and analysis of complete systems, including e.g. energy systems and complex measurement systems incorporating a multitude of sensors and processes. A central aspect in his research is to understand and model uncertainties and uncertainty propagation on multiple levels, to ensure that solutions and concepts are robust. He has also has extensive experience in managing research projects.
Kocbach has a Dr. Scient. (PhD) degree in Physics from the University of Bergen (2000). The PhD project involved design of ultrasonic piezoelectric transducers, including development of a finite element code (FEMP, still in active use at several entities including UiB) for piezoelectric transducers with propagation of acoustic waves into fluid media. From 2000 to 2005 he worked in the research and technology department in Nera with development of electromagnetic technology. Knowledge on sensor technology has been further extended to multiphysics systems, flow measurements, guided waves and other sensor principles through the work at Christian Michelsen Research since 2005 (NORCE since 2018 through merger).
System level application areas include a wide range of areas, including whole building energy systems, geothermal energy systems, flow assurance applications, measurement stations along the hydrogen supply chain, complex measurement systems within the O&G sector and measurement systems for monitoring fish farm nets.
Kocbach has been engaged in scientific programming for over 25 years, proficient in MATLAB, Python, and web programming. He is responsible for developing various in-house simulation models and tools like FEMP and VKBsim, and has a broad experience with commercial simulation tools like COMSOL Multiphysics, both used directly and via APIs. Data analysis and signal processing approaches include both traditional mathematical and statistical methods and machine learning. His work is characterized by focus on streamlining processes through automation, enhancing efficiency and productivity.
